chapter  6
10 Pages

Reliability

In the previous chapter, we discussed issues pertaining to measurement error in the behavioral and social sciences. However, none of our discussion was focused on means of quantifying its impact on observed scores. We will address this important problem in the present chapter that draws substantially on classical test theory (CTT) as well as the developments and notation introduced in Chapter 5. Before we commence, we note that the entire discussion in the present chapter will be at the population level, whereas matters pertaining to estimation of involved parameters will be addressed in the next chapter.